Frequently Asked Questions about the Speechly SLU platform and API.
SLU rules define the SLU model. The model is configured through sample utterances that are annotated using our custom syntax.
Utterance is something that the end user says. Utterance can consists of one or more segments.
It depends fully on the complexity of the model. If your model is trained to only turn lights on and off, probably 50 lines of utterances is already a pretty good amount. For any more complex application the amount of sample utterances should rather be in the thousands.
Application ID is needed to build an access token for the Speechly API. You can find your application ID in the Speechly Dashboard. Application ID should not be shared publicly, because it allows others to access your model. If you fear your application ID has leaked, you can either delete the application or ask us to revoke access to it.
You can test your model in the Speechly Playground. You can share access to your model in the Speechly Dashboard.
It depends heavily on the amount of SLU rules. It shouldn’t take more than 5 minutes with most setups, but if you have thousands of lines of rules, it can take even more. You will get a notification once the training is completed.
You can do this easily in the Playground window pushing the Share-button on the top right corner. When sharing the model you are asked to set the visibility of your app to Public and describe the App purpose so the users know what is it intended to be used for.
The Speechly API can be called from whatever device or platform. To do this you please review the API Reference to see how to call the API and received responses. We have also created ready made Libraries / SDKs that help you connect clients on the most popular platforms to the API. You can find documentatio for client libraries here.
We will launch our more advanced analytics features shortly, but in the meanwhile you can test your model in the Playground. A scientific way to evaluate the performance of the SLU models could be to have a separate set of text utterances and test the performance of your model in Playground with the same set of utterances.
You can delete all user data from the Admin Dashboard. If you want to verify all data is deleted, you can contact firstname.lastname@example.org
The easiest way to start developing on Speechly is our Quick Start tutorial for Speechly Web Client library. You can also revisit our simple NodeJS client in GitHub. We offer client libraries for iOS and Android, too. Contact us for an access to those libraries.
You can either use our API reference or contact us
Device ID is the unique identifier for the end user device. It is used in our API to differentiate between the different devices using the same model. The application id is used to adapt the speech recognition to the particular acoustic properties of the microphone of the device, the speaker and the usage context of the application. Keeping this ID persistent and distinct for each user/device will greatly improve the accuracy of the voice user experience.
The Speechly SLU model is based on machine learning and if your intents and entities are not recognized correctly, we recommend you to add more training rules that contain those intents and entitites to the configuration and trying again.
Please send us an email and we can help you forward.
Last updated by ottomatias on March 26, 2020 at 16:45 +0200
Found an error on our documentation? Please file an issue or make a pull request